The ability of the retina to detect the direction of motion has puzzled visual neuroscientists for over forty years. Specialized direction selective (DS) retinal ganglion cells respond to light in one (preferred) direction, but have little or no response in the opposite (null) direction. This direction selectivity is, in part, produced by starburst amacrine cells (SACs) through their release of the inhibitory neurotransmitter GABA. Physiological data demonstrates an asymmetric inhibitory input from SACs to DS cells; however the synaptic connections underlying directional selectivity are not understood. Previous anatomical studies have focused on potential contacts between these two cell types rather than actual synapses. The overall goal of the current proposal is determine the anatomical basis underlying direction selectivity within the retina. Specific Aim 1 will test the hypothesis that SAC varicosities are synaptic contacts. We will look for pre-synaptic markers for both GABA and acetylcholine in the varicosities, apposed by a post-synaptic GABAA receptor. This approach will allow us to detect synaptic structures using the confocal microscope. Specific Aim 2 will test the hypothesis that in the rabbit retina, SACs originating from the null side make more synaptic contacts onto DS cells than do the SACs arising from the preferred side. We will record from ON/OFF DS cells to establish the null/preferred axis and then dye fill the ganglion cell and a SAC on each side. Using the pre- and post-synaptic markers developed in Aim 1, we will use triple label confocal microscopy to test the hypothesis that there are asymmetric directionally selective synaptic contacts on the null side. Specific Aim 3 will use a recently developed double transgenic mouse line where nasal preferring DS (nDS) ganglion cells express GFP and the SACs express td-tomato. Because the nDS-GFP ganglion cells have a known preferred direction, we can omit the physiological experiments to determine the null/preferred axis. By filling the DS ganglion cell and SACs on both the null and preferred sides, combined with the pre- and post-synaptic markers above, we will test the hypothesis that there are significantly more synapses between null-side SACs and nDS cells, than preferred-side SACs and nDS cells. These data will demonstrate for the first time the anatomical synaptic connectivity responsible for the detection of motion within the retina. Furthermore, understanding of the complex neuronal circuitry of ganglion cells employed in the detection of motion may be instrumental in developing new innovative approaches to restoring vision following traumatic injury to the eye or disease.